TennisEdge's AI model finds no betting value in D. Shapovalov vs R. Hijikata (Los Cabos on 2026-07-27) — the market is efficient, with D. Shapovalov favoured at 1.80.
D. Shapovalov (A, rank 37, age 27) vs R. Hijikata (B, rank 96, age 25) — Los Cabos, HARD, 1/16-finals
═══ MODELS vs MARKET ═══
ELO A 0.577 | ML A 0.619 | Sim (fresh resim) A 0.6114 | Exchange A 1.65→0.606 | Book A 1.80
The brief's model_prob_a (0.521) was a STALE outlier; the authoritative re-simulation
returns A 0.6114, which converges tightly with ELO, ML and the exchange. Four-way
consensus: Shapovalov ~59-61%. ELO(0.577)≈rank(37) and Hijikata ELO-dog≈rank(96) — both
credible, no ELO/ranking divergence to exploit.
═══ RECENT FORM (all grass/clay; match is HARD) ═══
Shapovalov (1-5 window, FADING): L Carreno-Busta 3-6 6-7 (last match), L De Minaur 6-4 6-1,
W Pinnington-Jones 4-6 6-4 7-6 (GRIND, only win), L Cilic (3s), L Faria (straights).
Poor recent shape — but the slump is on clay/grass, his weaker surfaces; hard court is his
best surface (supports the favourite read, i.e. supports the sim/ELO/ML, not a fade).
Hijikata (mixed→FADING): L De Jong 7-6 3-6 5-7 6-4 6-3 (5-set loss to a 1474, last match),
L Humbert 6-1 6-2 (blown out), W Lehecka 1819 (GRIND over), W Tabilo (DOM), W London
Final vs Giron ~43d ago (grass title, too old + wrong surface to drive hard-court form).
═══ PATH / RUST / TRANSITION ═══
Both last played ~Jun 29-30 (Wimbledon) → ~28-day layoff = RUST RISK for BOTH. Neither is
match-sharp (first match at Los Cabos). Both TRANSITION grass→hard: depressed level and
WIDER margin variance for both — cuts both ways, supports no cover.
═══ FATIGUE ═══
Both well_rested, 0 matches/14d. No fatigue edge; this is a rust spot, not a fatigue spot.
═══ MARKET EDGES (commission-adjusted @ executable exchange prices) ═══
ML Shapov 1.65: sim 0.611 vs implied 0.618 → -0.7% (also below min_odds 1.717 = auto-PASS)
ML Hijik 2.46: sim 0.389 vs implied 0.419 → -3.0%
AH Shapov -1.5 @1.79: covers 0.583 vs 0.571 → +1.1% (below 5% bar)
AH Hijik +1.5 @1.97: covers 0.417 vs 0.520 → -10.3%
TOT Over 22 @1.79: 0.484 vs 0.571 → -8.7% | Under 22 @1.87: 0.516 vs 0.548 → -3.2%
Sim inputs (Shapov spw .612/rpw .409; Hijik spw .586/rpw .415, 50 matches each) are
coherent and already converge with ELO/ML/exchange — no justified override, baseline trusted.
═══ GATES ═══
G1 edge-origin: FAIL — best numerical edge is AH Shapov -1.5 at +1.1%, below the 5% bar;
every other market is negative at executable prices. No edge to build a bet on.
G2 consensus: FAIL — ELO+ML+sim+exchange all agree Shapovalov ~60%; no independent
evidence (H2H / live-event) contradicts the consensus.
G3 fatal-risk: n/a (no live thesis) — but mutual rust + grass→hard transition widens
margin variance and would negate any cover thesis anyway.
═══ VERDICT: PASS ═══
Textbook efficient market: the exchange (1.65) prices Shapovalov exactly where the fresh
sim, ELO and ML land. No side of ML, handicap or totals clears its bar. Both players rusty
and transitioning surfaces — high uncertainty on top of zero edge. No bet.